2026-11-02 ~ 11-08 · Generative
Generative · Personalization · AI Search · Infra · Embodied AI
- 1. PFA: Attention Can Be Progressively Focused in Long-Sequence Generative Recommendation · 📖 Deep dive
When a generative recommender decodes with beam search, every beam at every step computes attention over the whole user history, so longer histories mean slower serving. The authors at Alibaba found that attention…
Appeared in: 11-02
- 2. Towards Reliable, Scalable, and Value-Aligned Generative Recommendation for Large-Scale Advertising · 📖 Deep dive
Generative retrieval for ads has a basic mismatch: the model picks ads by generation probability, while the platform cares about revenue. Kuaishou measured it. The top ten ads by generation probability and the top ten…
Appeared in: 11-02
- 3. UGR: Unified Generative Ranking for Heterogeneous E-commerce Recommendation
Generative ranking in e-commerce recommendation has a mismatch: the list the model learns to generate is not the list users finally see. One screen mixes organic content, ads and items from other businesses…
Appeared in: 11-02
- 4. RoFiGR: Coarse Routing, Fine Refinement via Level-Adaptive Distillation in Multimodal Generative Recommendation · 📖 Deep dive
A generative recommender produces an item's SID one code at a time, and the first codes set the broad direction. One wrong branch early on and no amount of later precision can recover the relevant items…
Appeared in: 11-02
- 5. R2QRec: Learning Recommendation Queries from Retrieval Feedback
One way to use an LLM for recommendation is to have it write a few search queries from the user's history and pass them to an existing retrieval system. The catch is that logs record what the user clicked…
Appeared in: 11-02
- 6. Statistical Detection of Positional Bias in LLM-Based Recommender Systems: Methodologies and Empirical Evidence
Put candidate products into a prompt and ask an LLM to recommend one, and the order of the candidates by itself changes the pick. The authors built a RAG recommendation testbed on a real product catalog…
Appeared in: 11-02
- 7. Merge and Branch: Efficient Tree-Structured Supervised Fine-Tuning of LLMs for Recommendation
In LLM-based generative recommendation, the input is a long, text-heavy history while the output is an SID of just a few tokens. Training spends a lot of compute reading a whole history and then usually learns only one…
Appeared in: 11-02
- 8. Federated Recommender Agents
LLM recommender agents face a choice: pooling every user's memories raises privacy concerns, while isolated agents cannot learn from each other. FedRecAgent lets each client-side agent upload only experiences written…
Appeared in: 11-02
- 9. PALATE: Pretrained Autoregressive LATent Embeddings
Ads ranking on food-delivery and local-commerce marketplaces needs strong personalization, yet user histories are sparse and there are hundreds of thousands of stores. DoorDash's ranker used to rely on hand-built count…
Appeared in: 11-02
- 10. Category-Anchored Large Language Models for Next POI Recommendation
When predicting a user's next point of interest (POI), asking an LLM to generate the POI ID directly is unreliable: POIs are numerous, scattered and long-tailed, and their names carry little meaning…
Appeared in: 11-02